Disparate Discipline Insanity

On July 26th, Obama issued an executive order which received much less media attention than I expected it would.


After the Daily Caller website reported it, no one dared to touch it, but it’s insane. Here’s the beginning of the article:

President Barack Obama is backing a controversial campaign by progressives to regulate schools’ disciplinary actions so that members of major racial and ethnic groups are penalized at equal rates, regardless of individuals’ behavior.

His July 26 executive order established a government panel to promote “a positive school climate that does not rely on methods that result in disparate use of disciplinary tools.”

“African Americans lack equal access to highly effective teachers and principals, safe schools, and challenging college-preparatory classes, and they disproportionately experience school discipline,” said the order, titled “White House Initiative On Educational Excellence.”

I won’t even bother detailing how crazy it is, except to mention that the predictable consequence is even more misbehavior by black students and greatly accelerated “white flight” from the public schools.

It was bad enough when tests which blacks couldn’t pass at the same rate whites could were deemed unfair and racist and biased on that basis alone. But this policy assumes not that all races are equal in every intellectual area, but that they behave equally too. The obvious next step is a program to free black men from prison so the prison population will have the correct proportion of each race. Of course they may be able to adjust the black-white proportion to match the general population, but they’re going to have to start arresting Asians and jailing them on trumped up charges to achieve true racial equality. Maybe male prisoners should start suing because the laws defining felonies disparately impact them compared to women.

I wondered whether Obama really believes this shit, or whether he was just seeking to give advantages to blacks by any means necessary. The problem is a purely mathematical one: anything you do to equalize the proportion of students being suspended or expelled from school who are black to the proportion of the overall student population who are black can only achieve equality of outcome by inequality of treatment — white students will start suing because if you look at the collection of students who commit a given offense, one race will be punished much less than another. In other words, you must have one kind of disparate impact or another.

In my opinion, Obama is not quite THAT stupid. Therefore, he knows that this will result in more black misbehavior due to reduced punishments, and harsher treatment of white students, and he is fine with that.

But since this is about race, he is confident no one will challenge him because they are all terrified of being accused of being “racist”. I can’t see why this wouldn’t be a great issue to run on, but Romney is focusing on the economy. SOMEBODY needs to challenge it.

I have been planning to talk to my local school board about this and try to get a statement from them declaring that they would not adopt a policy that attempted to achieve racial equality in discipline outcomes by imposing racial inequality in the punishment of infractions. But since no one had picked this story up, I figured I had better wait until it looked like steps were actually being taken to implement the policy.

I didn’t have to wait long.

The Southern Poverty Law Center, a bunch of scam artists whose main activity is exaggerating the threat of white racism in order to offer a feeling of moral superiority to those liberals who donate to them, apparently got a heads-up and filed a lawsuit last month against the school district of Okaloosa County, Florida.

The SPLC filing is here:


and it is full of gaps in logic, arguments that school principals in the district are biased against black students based on nothing other than the statistical result that a greater proportion of suspended or expelled students are black (24%) than the proportion of students in the district as a whole are black (12%).

It seemed like this would get thrown out of court unless there was some statistical science to back up the hypothesis that the disparity was due to bias rather than actual differences in misbehavior. On the last page of the filing, the SPLC finally makes a scientific citation: a study published in 2000, called “The Color of Discipline: Sources of Racial and Gender Disproportionality in School Punishment”, lead author Russell Skiba of Indiana University.

Since I am a professional statistician, I read the paper carefully, and I can affirm that it is one of the most deceptive and tendentious academic papers I have ever reviewed. It is conclusory but never proves its contentions, and when you hack away all the obfuscatory verbiage, and draw inferences from the data they published about the numbers they egregiously left out, the central technical result of the paper is simply that black and white students tend to be punished for different kinds of offenses, and the offenses blacks are punished for tend to be more “subjective” (meaning that IF a principal were biased against black students then he would have the OPPORTUNITY to impose disparate punishments by using subjective standards).

But that merely provides a possible mechanism for how a bias could result in disparate punishments, this is NOT evidence that a bias exists. It merely shows that the bias explanation of the racially disparate outcomes is not inconceivable, but the simpler explanation is that the punishments reflect the actual offenses. To claim that bias is proven by this is to beg the question.

The data that is egregiously missing from the paper is the actual frequency of the offenses for each race. Reading between the lines, and interpreting the various complex (and intimidating to non-mathematicians) statistical coefficients that they do give, I surmise that black students commit most infractions at a higher rate than white students do, but that the ratio of the rates is higher for certain infractions labeled as “subjective” by the authors. In any case the absence of data is suspicious.

Another confusion made in the paper and repeated by the SPLC filing is the following two-step dance

1) In your literature review, carefully overlook research that leads to the opposite conclusion than yours

2) In your argument, use slippery language that equivocates between “absence of evidence” and “evidence of absence” and gives the impression that there were studies which found that actual misbehavior was equally distributed by race, rather than that you failed to discover studies showing the opposite.


About Polymath

Discoverable with effort
This entry was posted in Uncategorized. Bookmark the permalink.

One Response to Disparate Discipline Insanity

  1. Polymath says:

    This is happening in NYC.

    “If the Bloomberg administration is serious about helping young men of color succeed, then it must address these disparities and focus more resources on educating children, not arresting them,” argued Donna Lieberman, executive director of the New York Civil Liberties Union.

    The NYPD is pushing back hard:

    Top NYPD spokesman Paul Browne said Lieberman was ignoring the circumstances of the complaints that prompted the arrests.

    “The NYCLU’s kneejerk reaction to claim racism is as old as it is false,” he said. “It knows better than to compare arrests against population instead of description of suspects provided by victims.”

    I’m looking forward to hearing what Bloomberg says.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s